𝒍𝒆𝒎𝒂𝒏𝒏 @ lemann @lemmy.dbzer0.com

Hey 👋 I'm Lemann: mark II

---

Read more

Posts

6

Comments

486

Joined

2 yr. ago

I would suggest Plex HTPC on the deck instead, since it works really well with the deck's gamepad in the normal gaming mode. Touch also works but should probably be considered a last resort IMO

I personally think some types of openly developed software projects should have a strict non-commercial license: if companies aren't willing to contribute back to the source IMO they shouldn't be granted permission to freeload & have volunteers fix issues their paying customers run into

Donations are possibly a bit of an exception here - there are quite a few companies that still do this, albeit growing slimmer by the day.

Another big problem IMO is the subset of users that start attacking maintainers and volunteers because their "free app stopped working" etc. I see that a lot, mostly in the arduino community, but especially egregiously on the Zabbix project - I imagine a lot of those users are companies who aren't even paying/donating to the project

The hospitals in my nearby city have their own BRT which is open to public use, and joined to the city's ticketing system. It shuttles between them and various key locations, and is of course wholly subsidized for the intended users.

Despite being the only BRT here it pretty much goes everywhere it should, skipping the usual traffic, and as a result gets a lot of use.

If the users were limited to the regular transportation I think they would just all drive - while there are a lot of routes here they're not entirely pleasant to use IMO and almost always get stuck in traffic

By making something private, IMO you're revoking its status as an inherent right, and restricting access to only those who can afford it.

IMO the primary issue with this specifically venture funded stuff is that it'll undercut the competition for as long as possible, and then hike the prices once the competition has ceased to exist. When other recurring sources of funding are available, as it usually is with public services, this kind of thing is much less of an issue

I have a steam deck, and while its somewhat portable, I wouldn't want to carry one just to use at a charging station, considering the alternative is a controller stashed away in the glovebox. Probably also would want to carry the deck whenever leaving the vehicle too, considering how common it is in certain regions for some tesla models to be broken into and boot emptied out (apparently preventable with a 3d printed shim).

With the crazy powerful GPUs in these cars IMO it's practically free to enable this functionality for the user when the car is stationary, on a screen that is much, much larger and with a nicer sound system compared to any handheld or phone

IMO more accurate presence detection. Common sensors like PIR and cheaper doppler radar types can detect when there's motion, but not if a user is present but not moving in the detection area (e.g. sleeping or sitting). There's also open source projects that can track bluetooth wearables & phones to know who specifically is in a room, but these aren't able to detect people with no devices, say guests and kids.

The preferred approach at the moment is a combination of sensors to cover motion and person detection separately, which comes at a cost, both on your pocket and the time needed to get it working suitably for your needs, or maybe one of the more expensive radar sensors like the FP2 that can detect where in a space is occupied with higher accuracy than more affordable alternatives

The thermal cam is roughly in the same price range as the FP2 - however since it has I2C, something cheap like an ESP8266 can be used to turn it into a WiFi based presence detection sensor. Something like an ESP32 could be used to turn it into a presence detector and wearable tracker, negating the need for another separate sensor entirely.

Something like this would probably be quite close to plug and play for someone DIY focused, and wouldn't have the same problem as radar being able to see through walls to different areas, although this is somewhat solved by surrounding the rear of the sensor in foil with caveats.

I'm interested in how it performs outdoors in rain though, a lot of existing affordable sensors (except PIR) struggle a little in wet conditions, with doppler based ones not detecting anything

Yepp I agree lol, coverage about him/his companies gets old really fast...

Even worse when its stuff like laying off the one of his companies' most profitable business units over a childish disagreement 🤦‍♂️ just to reverse the decision after reporters start spamming his antics everywhere, leading to more spam.

Feel like he likes to see his name in the news or something, very tiring IMO

The !electronics@discuss.tchncs.de community would be a better fit for this post IMO

That said this looks interesting, and seems really competitively priced when compared to other similar thermal cameras like the Flir One Gen 3. This measures a very wide temperature range too, from -20°C to 400°C within ±2°C, whereas the Gen 3 tops out at 120°C.

The 80x62 resolution is OK for this application IMO, the Gen 3 is similar at 80x60.

I wouldn't go recommending one of these to a "casual" user though, it's very much a development product. Non-casual users could probably 3d print a case and get on just fine with the demo apps IMO, and for those users this would probably worth its weight in gold, given how expensive thermal imaging usually is.

Given how closely its priced to radar presence sensors like the Aqara FP2, and the inclusion of I2C for comms, I think this is going to have some appeal to DIY home automation enthusiasts too, especially if it gets supported by open source projects like ESPHome

Technical docs: https://www.waveshare.com/wiki/Thermal_Camera_HAT

Edit: I should also mention that there are much cheaper options on sites like AliExpress if you just want something cheap to use via usb-c - the advantages this sensor has are heavily dev focused IMO and might not make sense for most users (e.g. documentation, i2c comms, product integration support)

On models without a charge port, some of the batteries aren't even flat when the pods are burnt out 🤷‍♂️

Worst i've read is around 3.8v under load, which is roughly 60% full... that would be a very destructive result compressed in the back of a garbage truck.

Been harvesting these so far but don't have enough projects to use them in.

I had a similar stringing issue with PETG on my flsun Q5, where PLA was printing flawless without any strings whatsoever. Since your SR isn't a bowden, that mostly rules out retraction distance IMO.

For me I narrowed it down to Z-hop: as the printer was lifting the nozzle slightly to move to another part of the print, molten filament oozed out - sticking to the print, and getting stringed across by the nozzle. I only noticed after recording the print (use at least 60fps to avoid the frame blurring you'll get at 30> and watching what exactly was happening prior to the stringing. It took about 4 hours to get to that point 😭

If orcaslicer is based on Prusaslicer, wipe on retract with the initial distance set to 100% may not trigger a wipe. Try 70% or 80% instead. There's another setting somewhere to force retraction when changing layers: enabling this will forcibly honor your wipe retraction preference when changing print layer.

I think you may still get stringing at the really spiky parts of the test print, given there's no space for a wipe, and that volcano is going to hold way more molten filament than a V6 or similar

Hopefully someone with more experience can add their 2¢, it was pretty difficult to research info online when I had PETG stringing - everything just says "tune your retraction" 😤

Edit: I mixed up the SR with the v400! In that case, Another thing to check is your bowden tube pneumatic couplers - unload the filament and tug lightly on each to see if they're fitting securely. If they're loose, your extruder can end up moving the bowden tube itself back and forwards, which affects your retraction

shudders thinking about SAO

Not FOSS as it's under another license, but there's "FUTO Voice Input" if you're looking for a local alternative to Google's voice dictation on Android

https://gitlab.futo.org/alex/voiceinput

The repo has a list of supported and unsupported Android keyboards. Under the hood it uses OpenAI Whisper

I don't have hands-on experience with any of the brands listed so sadly can't give specific recommendations

Would a second-hand Google Pixel work fine for you, or do you definitely want to buy new? Mobile devices are really reliable nowadays, I doubt you'll need to worry about a warranty (which would not cover accidental damage, like a cracked screen)

I've noticed most countries outside of the US and Europe (making that assumption based on inability to get a Google Pixel or a Fairphone) typically have some locally used apps preinstalled regardless; usually a social networking and mobile e-payment app. I'd expect it to be possible to remove these using ADB if you have access to a computer.

Are you buying direct from manufacturer, or via a carrier? Do note that carriers may have customized the apps installed on your device if you buy one through them. For the least preinstalled bloat, buying direct is ideal

Link should be opened as if you were accessing it via dbzer0

Originally I watched the ad and couldn't see what was wrong with it (from the perspective of a relatively isolated software developer).

However the top comment on the Verge summed it up in a way I understood, making comparisons to the crushing of various creative devices in the ad (instruments, cameras, paints etc) and the current creative landscape in reality (game studios, music production etc).

With that in mind, the timing is pretty poor IMO and feels quite insensitive to creative individuals

I wouldn't be surprised if they also held the opinion that all PS3 games ran at 720p/30

A bunch of first party titles like Gran Turismo and Ratchet & Clank push native 1080p/60, and look pretty damn great especially considering the time period and 512MB (256MB VRAM) limitation.

Install steam. Run in big picture mode. Done. That's a steam machine. I don't get what you think a dedicated machine is going to do any differently. There is a reason Steam abandoned the idea themselves.

Big picture mode on my windows PC and the gamescope-focused UI on the Deck look similar, but offer very different capabilities IME.

To name a handful: FSR support for all games - including those that don't support it, per-game hardware performance profiles, excellent hardware integration - not just limited to the instant sleep and instant wake. With the third party Decky Store you can also configure the fan profile to your liking, control music apps running in the background on the Deck, and more. On the PC BPM these sadly do not exist

I 100% prefer playing on the deck any day of the week - the OS simply makes it so straightforward to jump into a game and forget about needing to also think about maintaining a desktop: no Windows updates, no telemetry service CPU spiking, and no Windows resetting my customized settings or forcing Edge browser defaults after an update.

That said, I don't particularly have an interest in a full blown Steam Machine - for me the Deck works just fine when docked.

I really want an updated steam controller with the same haptic touchpad tech present on the deck. The original controller, while comfortable, just doesn't compare to the improvements present on the Deck 😭

If the Humane could recognise speech on-device, and didn't require its own data plan, I'd be reasonably interested, since I don't really like using my phone for structuring my day.

I'd like a wearable that I can brain dump to, quickly check things without needing to unlock my phone, and keep on top of schedule. Sadly for me it looks like I'll need to go the DIY route with an esp32 board and an e-ink display, and drop any kind of stt + tts plans

Freetube, piped, invidious work great on desktop too